Position Title: Agentic AI Developer
Location: Irving-TX-USA, Alpharetta, GA
Work Arrangement: Onsite
Mode of Interview: Video /In person
Description:
***Locations: Irving-TX-USA, Alpharetta, GA
***Working Model: 100% Worksite based
We are seeking a skilled and forward-thinking Agentic AI Developer to design, build, and deploy autonomous AI agents and multi-agent systems. In this role, you will go beyond traditional static prompt engineering to create resilient, decision-making agentic workflows capable of planning, tool use, memory management, and autonomous execution to solve complex, multi-step real-world tasks.
You will collaborate closely with product managers, data engineers, and domain experts to take cutting-edge LLM agent architectures from research concepts into scalable, production-ready enterprise applications.
Key Responsibilities
Agent Design & Orchestration
- Architect and implement autonomous AI agents using modern frameworks (e.g., LangGraph, CrewAI, AutoGen, LlamaIndex, or custom orchestration loops).
- Design agent reasoning architectures, including ReAct loops, Plan-and-Solve patterns, Reflection/Self-Correction mechanisms, and hierarchical multi-agent structures.
- Construct robust tool-use pipelines enabling agents to seamlessly call external APIs, query databases, execute code, and interact with enterprise software.
System Architecture & Integration
- Integrate Long-Term Memory systems (vector databases, semantic search, key-value state stores) and Short-Term/Context Memory management strategies.
- Implement structured input/output parsing (Pydantic, JSON Schema) to guarantee reliable communication between model calls and execution environments.
- Build fallback mechanisms, human-in-the-loop (HITL) checkpoints, and deterministic guards around non-deterministic AI decisions.
Monitoring, Evaluation & Optimization
- Establish evaluation pipelines (using tools like Ragas, TruLens, or LangSmith) to benchmark agent accuracy, execution speed, trajectory quality, and tool failure rates.
- Monitor runtime performance, API usage costs, latency, and context-window token optimization across complex multi-step trajectories.
- Debug non-deterministic failures, looping behaviors, and hallucinations in multi-step agent execution paths.
Required Qualifications
- Education & Experience: Bachelor’s or Master’s degree in Computer Science, AI, Software Engineering, or equivalent practical experience with 1–2+ years focused on LLM/AI applications.
- Programming Skills: Expert-level proficiency in Python (TypeScript/JavaScript experience is a plus).
- Agent Frameworks: Demonstrated experience building with frameworks like LangGraph, CrewAI, Microsoft AutoGen, LlamaIndex Agents, or Semantic Kernel.
- API & Tool Integration: Proven track record building RESTful APIs, OpenAPI specifications, function calling architectures, and model tool definitions.
- Vector Databases & RAG: Hands-on experience with vector search engines (Pinecone, Qdrant, Milvus, Weaviate, or Chroma) and hybrid retriever architectures.
- Software Engineering Fundamentals: Strong background in asynchronous programming, microservices, Docker/Kubernetes, CI/CD, and unit/integration testing for non-deterministic AI systems.